Neo4j - Leveraging Chemical Reaction with Graphs - Knowledge Graphs

https://neo4j.com/news/the-chemical-knowledge-graph/ (very old tool these days...shows first efforts).
https://chemaxon.com/blog/presentation/similarity-graph-tool-interview
https://www.slideshare.net/neo4j/a-knowledge-graph-for-reaction-synthesis-prediction#1

https://github.com/neo4j/neo4j
https://neo4j.com/case-studies/novartis/?ref=web-solutions-life-sciences


Novartis Captures the Latest Biological Knowledge for Drug Discovery

The Challenge

Novartis has amassed decades of data on how various compounds affect protein targets, such as enzymes, with about a billion data points in all. That historical data is critical but sparse compared with the amazing granularity of the data currently being collected.

Today, Novartis uses an automated process that captures high-content image data showing how a particular compound has affected an entire cell culture. This generates terabytes of phenotypic data.

Novartis faced the challenge of combining its historical data stores with this burgeoning phenotypic data. They also needed a way to place all this data within the larger context of ongoing medical research from around the world.

The Novartis team wanted to combine its data with medical information from NIH’s PubMed. PubMed contains about 25 million abstracts from some 5,600 scientific journals.

The Novartis team sought a way to empower researchers to ask questions connecting the dots between all of this data in the context of the latest medical research.

As Stephan Reiling, Senior Scientist at Novartis, put it, “When we try to analyze this data, it becomes much more apparent that we need to have a way to store biological knowledge and then run queries against it.”

The Solution

Ingesting and connecting data about diseases, genes and compounds – along with identifying the nature of the relationships between these elements – held the promise of accelerating drug discovery.

The Novartis team wanted to link genes, diseases and compounds in a triangular pattern. “For successful drug discovery, you need to be able to navigate this triangle,” explained Reiling. The Novartis team decided to create a knowledge graph stored in Neo4j, and devised a processing pipeline for ingesting the latest medical research.

Text mining is used at the beginning of the pipeline to extract relevant text data from PubMed. That data is then fed into Neo4j, along with Novartis’s own historical and image data. The data pipeline populates the 15 kinds of nodes that were devised to encode the data. The next phase fills in the relationship information that links the nodes together. The team identified more than 90 different relationships.

Novartis uses Neo4j graph algorithms to traverse the graph and identify a desired triangular node pattern linking the three classes of data together. Graph analytics not only find relevant nodes in the desired triangular relationship, but also employ a metric the team designed to gauge the associated strength between each node in each triangle. Using this capability, the team devised queries to find data linked by the desired node pattern, with a given association strength, and then sort the triangles according to this metric.

When researchers query the knowledge graph, results show the strength of the correlation between elements. If a researcher already knows about a highly associated correlation, they might choose to investigate others, which could take their work in new directions.

https://go.neo4j.com/rs/710-RRC-335/images/Neo4j-case-study-Novartis-EN-US.pdf

https://neo4j.com/developer-blog/chemical-reactions-graph/
https://github.com/wagenrace/openmeasurement_database/blob/main/0001_populate_chemicals/README.md
https://www.slideshare.net/slideshow/chemical-similarity-using-multiterabyte-graph-databases-68-billion-nodes-and-counting/80690196#1

Graph using RDKit:

Global-Chem

https://github.com/Sulstice/global-chem

Organizing The Chemical Universe

Global Chem is a public dictionary of common chemical lists using the Common Chemical Name as input and SMILES/SMARTS as output organized by their respective community in a knowledge graph.

Global-Chem serves as an open source platform where writing the molecules directly allows for the name to not be ambiguous to what a chemical is is anymore and allows for full transparency.

Our hope is this repository serves as a base for the population to govern how the chemicals we use in things like Food, Clothing, Environment, Materials, Drugs, War and a lot more are beneficial for all of us.


https://www.youtube.com/watch?v=PSrrVOjnijA

This guy's github could be used with ChatGPT/Jupyter notebooks for creating a "Miles" style atom linkages using Cypher:

https://github.com/tomasonjo/blogs/blob/master/llm/generic_cypher_gpt4.ipynb

Also NeoDash's built in component:
https://neo4j.com/labs/neodash/2.4/user-guide/extensions/natural-language-queries/
https://towardsdatascience.com/cypher-generation-the-good-the-bad-and-the-messy-4ec119dd72ea
https://github.com/SolanaO/Blogs_Content/tree/master/cypher_generator

Website of Global Chem:


https://globalchemistry.org/

Article on chemical graph databases with Neo4j and modeling:

https://neo4j.com/developer-blog/kickstart-chemical-graph-database/
https://github.com/wagenrace/openmeasurement_database/tree/main/0001_populate_chemicals

Compound
This is related to a compound of PubChem. It has 1 property.

pubChemCompId: The id within pubchem. So “compound:cid162366967” links to https://pubchem.ncbi.nlm.nih.gov/compound/162366967.
This number can be used with both PubChem RDF and PUG.

https://neo4j.com/developer-blog/tagged/chemicals/
---------

RDKit Module:

https://github.com/rdkit/neo4j-rdkit

Abstract
Chemical and pharmaceutical R&D produce large amounts of data of completely different nature, such as chemical structures, recipe and process data, formulation data, and data from various application tests. Altogether these data rarely follow a schema. Consequently, relational data models and databases have frequetly disadvantages mapping these data appropriately. Here, chemical data frequently leads to rather abstract data models, which are difficult to develop, align, and maintain with the domain experts. Upon retrieval computationally expesive joins in not predetermined depths may cause issues.

Graph data models promise here advantages:

they can easily be understood by non IT experts from the research domains
due to their plasticity, they can easily be extended and refactored
graph databases such as neo4j are made for coping with arbitrary path lengths
Chemical data models usually require a database to be able to deal with chemical structures to be utilized for structure based queries to either identify records or as filtering criteria.

The project will be focused on development of extension for neo4j graph database for querying knowledge graphs storing molecular and chemical information. Task is to enable identification of entry points into the graph via exact/substructure/similarity searches (UC1). UC2 is closely related to UC1, but here the intention is to use chemical structures as limiting conditions in graph traversals originating from different entry points. Both use cases rely on the same integration of RDkit and Neo4j and will only differ in their CYPHER statements.